FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952826183> ?p ?o ?g. }

• W2952826183 abstract "When applying machine learning to sensitive data, one has to find a balance between accuracy, information security, and computational-complexity. Recent studies combined Homomorphic Encryption with neural networks to make inferences while protecting against information leakage. However, these methods are limited by the width and depth of neural networks that can be used (and hence the accuracy) and exhibit high latency even for relatively simple networks. In this study we provide two solutions that address these limitations. In the first solution, we present more than $10times$ improvement in latency and enable inference on wider networks compared to prior attempts with the same level of security. The improved performance is achieved by novel methods to represent the data during the computation. In the second solution, we apply the method of transfer learning to provide private inference services using deep networks with latency of $sim0.16$ seconds. We demonstrate the efficacy of our methods on several computer vision tasks." @default.
• W2952826183 created "2019-06-27" @default.
• W2952826183 creator A5015030979 @default.
• W2952826183 creator A5029118855 @default.
• W2952826183 creator A5036797884 @default.
• W2952826183 date "2018-12-27" @default.
• W2952826183 modified "2023-09-26" @default.
• W2952826183 title "Low Latency Privacy Preserving Inference" @default.
• W2952826183 cites W1522301498 @default.
• W2952826183 cites W1686810756 @default.
• W2952826183 cites W1992282993 @default.
• W2952826183 cites W2027471022 @default.
• W2952826183 cites W2031533839 @default.
• W2952826183 cites W2036329595 @default.
• W2952826183 cites W2092422002 @default.
• W2952826183 cites W2115733720 @default.
• W2952826183 cites W2149933564 @default.
• W2952826183 cites W2177209050 @default.
• W2952826183 cites W2194775991 @default.
• W2952826183 cites W2346062110 @default.
• W2952826183 cites W2435473771 @default.
• W2952826183 cites W2460519802 @default.
• W2952826183 cites W2768174108 @default.
• W2952826183 cites W2783769334 @default.
• W2952826183 cites W2788636062 @default.
• W2952826183 cites W2805074088 @default.
• W2952826183 cites W2805453438 @default.
• W2952826183 cites W2807403537 @default.
• W2952826183 cites W2898730285 @default.
• W2952826183 cites W2901469102 @default.
• W2952826183 cites W2914508875 @default.
• W2952826183 cites W2950438104 @default.
• W2952826183 cites W2950460048 @default.
• W2952826183 cites W2957780493 @default.
• W2952826183 cites W2962986039 @default.
• W2952826183 cites W2963106566 @default.
• W2952826183 cites W2977558206 @default.
• W2952826183 cites W3028867652 @default.
• W2952826183 cites W3118608800 @default.
• W2952826183 doi "https://doi.org/10.48550/arxiv.1812.10659" @default.
• W2952826183 hasPublicationYear "2018" @default.
• W2952826183 type Work @default.
• W2952826183 sameAs 2952826183 @default.
• W2952826183 citedByCount "3" @default.
• W2952826183 countsByYear W29528261832018 @default.
• W2952826183 countsByYear W29528261832019 @default.
• W2952826183 countsByYear W29528261832020 @default.
• W2952826183 crossrefType "posted-content" @default.
• W2952826183 hasAuthorship W2952826183A5015030979 @default.
• W2952826183 hasAuthorship W2952826183A5029118855 @default.
• W2952826183 hasAuthorship W2952826183A5036797884 @default.
• W2952826183 hasBestOaLocation W29528261831 @default.
• W2952826183 hasConcept C113775141 @default.
• W2952826183 hasConcept C11413529 @default.
• W2952826183 hasConcept C119857082 @default.
• W2952826183 hasConcept C148730421 @default.
• W2952826183 hasConcept C154945302 @default.
• W2952826183 hasConcept C158338273 @default.
• W2952826183 hasConcept C2776214188 @default.
• W2952826183 hasConcept C2779201187 @default.
• W2952826183 hasConcept C2984842247 @default.
• W2952826183 hasConcept C38652104 @default.
• W2952826183 hasConcept C41008148 @default.
• W2952826183 hasConcept C45374587 @default.
• W2952826183 hasConcept C50644808 @default.
• W2952826183 hasConcept C76155785 @default.
• W2952826183 hasConcept C80444323 @default.
• W2952826183 hasConcept C82876162 @default.
• W2952826183 hasConceptScore W2952826183C113775141 @default.
• W2952826183 hasConceptScore W2952826183C11413529 @default.
• W2952826183 hasConceptScore W2952826183C119857082 @default.
• W2952826183 hasConceptScore W2952826183C148730421 @default.
• W2952826183 hasConceptScore W2952826183C154945302 @default.
• W2952826183 hasConceptScore W2952826183C158338273 @default.
• W2952826183 hasConceptScore W2952826183C2776214188 @default.
• W2952826183 hasConceptScore W2952826183C2779201187 @default.
• W2952826183 hasConceptScore W2952826183C2984842247 @default.
• W2952826183 hasConceptScore W2952826183C38652104 @default.
• W2952826183 hasConceptScore W2952826183C41008148 @default.
• W2952826183 hasConceptScore W2952826183C45374587 @default.
• W2952826183 hasConceptScore W2952826183C50644808 @default.
• W2952826183 hasConceptScore W2952826183C76155785 @default.
• W2952826183 hasConceptScore W2952826183C80444323 @default.
• W2952826183 hasConceptScore W2952826183C82876162 @default.
• W2952826183 hasLocation W29528261831 @default.
• W2952826183 hasOpenAccess W2952826183 @default.
• W2952826183 hasPrimaryLocation W29528261831 @default.
• W2952826183 hasRelatedWork W2784537340 @default.
• W2952826183 hasRelatedWork W2906483705 @default.
• W2952826183 hasRelatedWork W2952826183 @default.
• W2952826183 hasRelatedWork W4213041209 @default.
• W2952826183 hasRelatedWork W4286981775 @default.
• W2952826183 hasRelatedWork W4297798330 @default.
• W2952826183 hasRelatedWork W4308654840 @default.
• W2952826183 hasRelatedWork W4310419184 @default.
• W2952826183 hasRelatedWork W4312312641 @default.
• W2952826183 hasRelatedWork W4318348488 @default.
• W2952826183 isParatext "false" @default.
• W2952826183 isRetracted "false" @default.
• W2952826183 magId "2952826183" @default.

• next